Sparse Generalized Linear Model with L0 Approximation for Feature Selection

An efficient procedure for feature selection for generalized linear models with L0 penalty, including linear, logistic, Poisson, gamma, inverse Gaussian regression. Adaptive ridge algorithms are used to fit the models.

CRAN RStudio mirror downloads

l0ara fits regularization for linear or generalized linear models with L0 penalty. Adaptive ridge algorithms are used to fit the models.


To install the latest version from github :


To install from CRAN:



l0ara 0.1.5

  • added standardize option in l0ara.R
  • fixed cran issues

l0ara 0.1.4

  • fixed typos and added links in documentation
  • changed subtitle name in ROC plot

l0ara 0.1.3

  • fixed typos in documentation
  • fixed bugs when calculating crose-validated error
  • added type="auc" and type="class" for cv.l0ara
  • added plot method for cv.l0ara

l0ara 0.1.2

  • initial release on CRAN

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.1.6 by Wenchuan Guo, 2 years ago

Browse source code at

Authors: Wenchuan Guo , Shujie Ma , Zhenqiu Liu

Documentation:   PDF Manual  

GPL-2 license

Imports Rcpp

Linking to Rcpp, RcppArmadillo

See at CRAN